- From: Geoffrey Sneddon <foolistbar@googlemail.com>
- Date: Sun, 28 Oct 2007 21:39:53 +0000
- To: Julian Reschke <julian.reschke@gmx.de>
- Cc: ietf-http-wg@w3.org
On 28 Oct 2007, at 20:53, Julian Reschke wrote: > Geoffrey Sneddon wrote: >>> The simple answer is: the double quotes are part of the entity >>> tag. So a response header such as >>> >>> ETag: x >>> >>> would simply be invalid and should be ignored. >> I am aware ? but how is the receiving end meant to deal with them? >> Is it meant to keep the quotation marks around any quoted-string, >> even when that therefore results in non-exist things like a >> character set called "UTF-8" (with quotes)? Or does the behaviour >> need to be specific to each and every use of quoted-string need to >> have it defined separately? > > That may be the case. I was referring to my original example, i.e., > If you don't [parse the quotation marks out], you can end up with > character sets such as "UTF-8" (i.e., including the quotation > marks) in headers like Content-Type: text/plain;charset="UTF-8". Which really is the question: what are we meant to do with the delimiting quotation marks in quoted-string? If we take UTF-8 as a string, we can escape this as a quoted-string in several ways, including: - "UTF-8" - "\U\T\F\-\8" Now, are we meant to unescape every quoted-string we come across (therefore including entity-tag), or only some? I think we can all agree that "\U\T\F\-\8" is not, in itself, a valid character set. If only some, which? As it stands now, it is not clear if you should ever unescape them. - Geoffrey.
Received on Sunday, 28 October 2007 21:40:13 UTC